#!/usr/bin/env python3
"""
Tier 1 Scorer - Full Database Scan
- Processes ALL leads JAE v5 has researched
- Updates CRM fields directly
- Creates filterable views
"""
import json, re, time, urllib.request, ssl
from datetime import datetime
from pathlib import Path

SCRIPT_DIR = Path(__file__).parent
STATE_FILE = SCRIPT_DIR / "state" / "tier1-state.json"
JAE_STATE = SCRIPT_DIR / "state" / "jae-v5-state.json"
CRM_URL = "https://salesforce.hoaledgeriq.com/rest"
CRM_TOKEN = "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiI5M2FmNGFmNS0zZWQ0LTQ1ZDMtOWE5Zi01MDMzZjc3YTY3MjMiLCJ0eXBlIjoiQVBJX0tFWSIsIndvcmtzcGFjZUlkIjoiOTNhZjRhZjUtM2VkNC00NWQzLTlhOWYtNTAzM2Y3N2E2NzIzIiwiaWF0IjoxNzczMzI4NDQzLCJleHAiOjE4MDQ3ODE2NDIsImp0aSI6IjIwZjEyYzkwLTRkMDctNGJmNi1iMzk3LTZjNmU3MzlmMThjOCJ9.zeM5NvwCSGEcz99m2LYtgb0sVD6WUXcCF7SwonFg930"

ssl_context = ssl.create_default_context()
ssl_context.check_hostname = False
ssl_context.verify_mode = ssl.CERT_NONE

def log(msg):
    ts = datetime.now().strftime('%H:%M:%S')
    print(f"[{ts}] {msg}")

def fetch_all_notes_paginated():
    """Fetch all notes with pagination"""
    all_notes = []
    has_more = True
    end_cursor = None
    
    log("Fetching all leads from CRM (with pagination)...")
    
    while has_more:
        try:
            url = f"{CRM_URL}/notes?limit=200&order[createdAt]=desc"
            if end_cursor:
                url += f"&after={end_cursor}"
            
            req = urllib.request.Request(
                url,
                headers={"Authorization": f"Bearer {CRM_TOKEN}", "Accept": "application/json"}
            )
            opener = urllib.request.build_opener(urllib.request.HTTPSHandler(context=ssl_context))
            with opener.open(req, timeout=30) as r:
                data = json.loads(r.read().decode())
                notes = data.get('data', {}).get('notes', [])
                all_notes.extend(notes)
                
                # Check pagination
                page_info = data.get('pageInfo', {})
                has_more = page_info.get('hasNextPage', False)
                end_cursor = page_info.get('endCursor')
                
                log(f"  Fetched {len(notes)} leads (total: {len(all_notes)})")
                
                if not has_more:
                    break
                    
        except Exception as e:
            log(f"Fetch error: {e}")
            break
    
    log(f"Total leads fetched: {len(all_notes)}")
    return all_notes

def extract_units_from_note(note):
    """Extract unit count from note body or title"""
    body = note.get('bodyV2', {}).get('markdown', '') if isinstance(note.get('bodyV2'), dict) else ''
    title = note.get('title', '')
    text = f"{title} {body}".lower()
    
    # Look for unit patterns
    patterns = [
        r'units:\s*(\d{1,4})',
        r'(\d{1,4})\s*(?:homes|units|lots|properties|residences)',
        r'community\s*of\s*(\d{1,4})',
        r'(\d{1,4})\s*home\s*owners',
    ]
    
    for pattern in patterns:
        match = re.search(pattern, text, re.IGNORECASE)
        if match:
            try:
                units = int(match.group(1))
                if 10 <= units <= 5000:
                    return units
            except:
                pass
    return None

def has_budget_pdf(note):
    """Check if note has budget PDF"""
    body = note.get('bodyV2', {}).get('markdown', '') if isinstance(note.get('bodyV2'), dict) else ''
    title = note.get('title', '')
    text = f"{title} {body}".lower()
    
    # Check for budget mentions (JAE v5 format)
    if 'budget pdf' in text or 'budget.pdf' in text or 'found budget pdf' in text:
        return True
    # Also check title patterns from JAE research
    if 'budget' in text and 'found' in text:
        return True
    return False

def has_website(note):
    """Check if note has website"""
    body = note.get('bodyV2', {}).get('markdown', '') if isinstance(note.get('bodyV2'), dict) else ''
    title = note.get('title', '')
    text = f"{title} {body}"
    return 'https://' in text or 'http://' in text

def get_temp(note):
    """Get temperature from note"""
    temp = note.get('temp', 'COLD')
    if temp and temp.upper() in ['HOT', 'WARM', 'COLD']:
        return temp.upper()
    
    title = note.get('title', '').upper()
    if title.startswith('HOT:'):
        return 'HOT'
    if title.startswith('WARM:'):
        return 'WARM'
    return 'COLD'

def score_lead(note):
    """Score a lead 1-10 based on Tier 1 criteria"""
    temp = get_temp(note)
    units = extract_units_from_note(note)
    budget_pdf = has_budget_pdf(note)
    has_site = has_website(note)
    
    score = 0
    
    # 1. Temperature (max 3 points)
    if temp == 'HOT':
        score += 3
    elif temp == 'WARM':
        score += 2
    
    # 2. Unit Count (max 4 points)
    if units:
        if 150 <= units <= 400:
            score += 4
        elif 100 <= units < 150 or 400 < units <= 500:
            score += 3
        elif 50 <= units < 100 or 500 < units <= 1000:
            score += 2
        else:
            score += 1
    
    # 3. Budget PDF (max 2 points)
    if budget_pdf:
        score += 2
    elif has_site:
        score += 1
    
    # 4. Website (max 1 point)
    if has_site:
        score += 1
    
    return score

def update_crm_note(note_id, score, tier1_label):
    """Update CRM note with tier1_score field"""
    try:
        patch_data = json.dumps({
            "tier1Score": score,
            "tier1Label": tier1_label
        }).encode()
        
        req = urllib.request.Request(
            f"{CRM_URL}/notes/{note_id}",
            data=patch_data,
            headers={
                "Authorization": f"Bearer {CRM_TOKEN}",
                "Content-Type": "application/json"
            },
            method='PATCH'
        )
        
        opener = urllib.request.build_opener(urllib.request.HTTPSHandler(context=ssl_context))
        with opener.open(req, timeout=20) as r:
            return True
    except Exception as e:
        return False

def main():
    log("=" * 60)
    log("Tier 1 Scorer - Full Database Scan")
    log("=" * 60)
    
    # Fetch all notes
    notes = fetch_all_notes_paginated()
    
    total_scored = 0
    crm_updates = 0
    tier1_count = 0
    
    for i, note in enumerate(notes):
        note_id = note.get('id')
        title = note.get('title', '')[:50]
        
        # Score the lead
        score = score_lead(note)
        
        # Determine Tier 1 label
        if score >= 8:
            tier1_label = "Tier 1 - Priority"
        elif score >= 6:
            tier1_label = "Tier 1"
        else:
            tier1_label = ""
        
        # Update CRM if score is 6+
        if score >= 6:
            total_scored += 1
            if update_crm_note(note_id, score, tier1_label):
                crm_updates += 1
                tier1_count += 1
                if tier1_count <= 10:  # Show first 10
                    log(f"  ✓ {title[:40]} (Score: {score}/10)")
        
        # Progress indicator
        if (i + 1) % 100 == 0:
            log(f"Processed {i+1}/{len(notes)} leads...")
    
    log("\n" + "=" * 60)
    log(f"Tier 1 Scoring Complete!")
    log(f"  Total leads processed: {len(notes)}")
    log(f"  Tier 1 leads (6+): {tier1_count}")
    log(f"  CRM updates: {crm_updates}")
    log(f"\n📊 Filter in CRM:")
    log(f"  • View: tier1_score >= 6")
    log(f"  • Sort by: tier1_score DESC")
    log("=" * 60)

if __name__ == "__main__":
    main()